Algorithmic or automated underwriting (AU) has become increasingly popular in the mortgage markets. AU systems do not use race as an input and presumably base loan decisions only on the applicant's credit history and financial data. Because of the increasing popularity of AU, we need to study its impact on antidiscrimination efforts in the mortgage lending business. This article explores new data and statistical techniques, finding that AU seems to reduce but not eliminate discrimination in the mortgage markets. Policy makers may want to encourage the use of AU, but they shouldn't count on it to eliminate discrimination entirely.
This article appeared in the First Quarter 2022 issue of Economic Insights. Download and read the full issue.
